TPS54240: Reverse current protection

GeT

Intellectual 990 points

Part Number: TPS54240

Does the TPS54240 have reverse current protection?

If a DC voltage appears on its output (from another source) while the input is 0V, will be there any problem?

over 7 years ago

0 Samuel Jaffe over 7 years ago

TI__Mastermind 25590 points

GeT,

If VIN=0V and VOUT=something significant:

The body diode from the internal high-side FET will conduct current through to VIN. The part has current sense but that's for slope compensation and it's only active if the part is on.

You would want to add reverse current blocking. Or look into battery chargers. They are made to operate in this condition.
-Sam

0 GeT over 7 years ago in reply to Samuel Jaffe

Intellectual 990 points

Would the chip be damaged or destroyed by something like that?

0 Samuel Jaffe over 7 years ago in reply to GeT

TI__Mastermind 25590 points

GeT,

We do not test or guarantee operation in this condition.

That said, a small amount of current will probably be fine.

-Sam

0 GeT over 7 years ago in reply to Samuel Jaffe

Intellectual 990 points

There's no other circuits at the input of TPS54240, only the necessary passive parts to get it work (my design is exactly like the Figure 64 Page 40 in the datasheet). I suppose that the current would be small enough.

On normal operation the TPS54240 works with VOUT = 5V.

The voltage on its outpout in the case VIN = 0V, is 5V as well.

0 Samuel Jaffe over 7 years ago in reply to GeT

TI__Mastermind 25590 points

GeT,

Here's another consideration. If VOUT is high enough to charge VIN high enough to turn on the part, the part may start switching, turn itself off, then on, then off... but that depends on the load.

Is your load a battery? A voltage source? A charged cap?

-Sam

0 GeT over 7 years ago in reply to Samuel Jaffe

Intellectual 990 points

The part starts switching above 6.5V (through EN pin)and the VOUT is 5V comes from a USB port. (VUSB - VD ~= 4.7 to be exact)

VUSB is isolated by a TVS diode.

In normal operation, when VIN is > 6.5V, VOUT is 5V as well.

0 Samuel Jaffe over 7 years ago in reply to GeT

TI__Mastermind 25590 points

GeT,

Alright, talked with another pro in the office and our consensus is that you should avoid letting current go through the body diode. You should, instead, add a schottkey diode from VOUT to VIN so the current flows through the schottkey if/when VIN goes low.

If VOUT is 5V and EN is set up properly, you shouldn't have a problem letting 5V sit on VIN.

-Sam

0 GeT over 7 years ago in reply to Samuel Jaffe

Intellectual 990 points

My setup for EN pin is R1 = 100K and R2 = 22K. I suppose there's no any problem with this voltage. Is there?

0 Samuel Jaffe over 7 years ago in reply to GeT

TI__Mastermind 25590 points

GeT,

That looks great.

-Sam

0 GeT over 7 years ago in reply to Samuel Jaffe

Intellectual 990 points

Thank you samuel!

So, the problem is that the part gets enabled and goes into an on/off loop.
If the EN threshold is high enough, there's no any problem.

0 Samuel Jaffe over 7 years ago in reply to GeT

TI__Mastermind 25590 points

Thank YOU!

0 GeT over 7 years ago in reply to Samuel Jaffe

Intellectual 990 points

"If VOUT is 5V and EN is set up properly, you shouldn't have a problem letting 5V sit on VIN."

Samuel, did you mean: "shouldn't have a problem letting 5V sit on VOUT." ?
Because the voltage on VIN appears only when I plug the USB cable and the VOUT gets ~5V, (throught the body diode of the internal mosfet).

Sorry for bringing up the post but I'm trying to find out if I lost something.

0 Samuel Jaffe over 7 years ago in reply to GeT

TI__Mastermind 25590 points

GeT,

Not a problem. Yes, if you have USB 5V on VOUT, the body diode will conduct and put ~5V on VIN. The only issue would be if ~5V at VIN turned on the part, but EN is set up to keep the part off when VIN ~5V, hence you have no problem. You could also say "shouldn't have a problem letting 5V sit on VOUT" but I said VIN because it would be VIN which actually caused the issue (by way of VOUT).

All this to say, both are true and you're good either way.

-Sam

0 GeT over 7 years ago in reply to Samuel Jaffe

Intellectual 990 points

Thank you again for your prompt reply!

Power management

Power management forum

TPS54240: Reverse current protection